accepted manuscript - tara
TRANSCRIPT
Accepted Manuscript
Title: Assessing Neuronal Networks: UnderstandingAlzheimer’s Disease
Authors: Arun L.W. Bokde, Michael Ewers, Harald Hampel
PII: S0301-0082(09)00096-3DOI: doi:10.1016/j.pneurobio.2009.06.004Reference: PRONEU 958
To appear in: Progress in Neurobiology
Received date: 20-2-2009Accepted date: 19-6-2009
Please cite this article as: Bokde, A.L.W., Ewers, M., Hampel, H., Assessing NeuronalNetworks: Understanding Alzheimer’s Disease, Progress in Neurobiology (2008),doi:10.1016/j.pneurobio.2009.06.004
This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.
Page 1 of 50
Accep
ted
Man
uscr
ipt
1
Assessing Neuronal Networks: Understanding Alzheimer’s Disease
Arun LW Bokde PhD, Michael Ewers PhD, and Harald Hampel* MD, MSc
Discipline of Psychiatry, School of Medicine and Trinity College Institute of Neuroscience (TCIN), Laboratory of Neuroimaging & Biomarker Research, Trinity College Dublin, The Adelaide and Meath Hospital incorporating the National Children’s Hospital (AMiNCH), Dublin, Ireland
and
Dementia and Neuroimaging Research Section, Alzheimer Memorial Center and Geriatric Psychiatry Branch, Department of Psychiatry, Ludwig-Maximilian University, Munich, Germany
* Corresponding Author
Running Title: Connectivity in AD
Corresponding Author:Harald Hampel, MD, MScDiscipline of PsychiatrySchool of MedicineTrinity CollegeTrinity Centre for the Health SciencesThe Adelaide and Meath Hospitalincorporating The National Children’s Hospital (AMiNCH)Dublin 24 IrelandTel.: +353-1-896 3706Fax.: +353-1-896 1313Email: [email protected]
* Manuscript
Page 2 of 50
Accep
ted
Man
uscr
ipt
2
Abstract
Findings derived from neuroimaging of the structural and functional organization of
the human brain have led to the widely supported hypothesis that neuronal networks
of temporally coordinated brain activity across different regional brain structures
underpin cognitive function. Failure of integration within a network leads to cognitive
dysfunction. The current discussion on Alzheimer’s disease (AD) argues that it
presents in part a disconnection syndrome. Studies using functional magnetic
resonance imaging, positron emission tomography and electroencephalography
demonstrate that synchronicity of brain activity is altered in AD and correlates with
cognitive deficits. Moreover, recent advances in diffusion tensor imaging have made
it possible to track axonal projections across the brain, revealing substantial regional
impairment in fiber-tract integrity in AD. Accumulating evidence points towards a
network breakdown reflecting disconnection at both the structural and functional
system level. The exact relationship among these multiple mechanistic variables and
their contribution to cognitive alterations and ultimately decline is yet unknown.
Focused research efforts aimed at the integration of both function and structure hold
great promise not only in improving our understanding of cognition but also of its
characteristic progressive metamorphosis in complex chronic neurodegenerative
disorders such as AD.
Page 3 of 50
Accep
ted
Man
uscr
ipt
3
Table of Contents
1.0 Introduction
2.0 Neuropathology and its spread in the brain
3.0 fMRI and PET findings: revealing connectivity in functional brain networks
3.1 Cognitive Function Domain
3.2 Coherent Resting Networks in the Brain
3.3 Interaction among networks
3.4 Connectivity Dysfunction due to Changes in White Matter
4.0 Future Perspectives of Research in Connectivity in AD
5.0 Clinical Applications
6.0 Conclusions
Page 4 of 50
Accep
ted
Man
uscr
ipt
4
1.0 Introduction
Functional neuroimaging studies in humans and animals suggest that particular brain
regions are necessary for specific cognitive functions. The activations of different
brain regions, however, do not appear to occur independently from each other but may
occur in a sequential spatio-temporally ordered fashion (McIntosh et al., 1994;
Murphy et al., 1993). The involved regions integrate into a large-scale network which
forms the basis of cognition and closely relates to its complex underlying systemic
structural architecture (Horwitz et al., 2005; Lee et al., 2006; Luria, 1973; McIntosh,
2004; Rogers et al., 2007). For example, successful associative learning has been
shown to correlate with a change in the effective connectivity, i.e., the influence of
activation of one brain region onto another, within a specific neuronal network
(Buchel et al., 1999). From the neuroanatomical perspective, connectivity of brain
activity is predicted to be confined towards pathways of neuroanatomical connections
between specific brain regions (Greicius et al., 2008; Toosy et al., 2004). These
neuroanatomical constraints allow generating useful predictive models, specific
working hypotheses concerning the effect of localized lesions on specific network
functions in complex chronically progressive neurodegenerative system disorders
such as Alzheimer’s disease (AD).
Thus a failure of the regions of a network to interact at a high level of coordination
may underpin the cognitive disorders which are present in AD. The failure of
network function may be due to interaction failure among the regions of a network,
which is denoted the disconnection hypothesis. In other words, a disruption in the
temporal-spatially coordinated activity among different regions in the brain rather
than isolated changes in particular brain regions may underlie cognitive impairment in
Page 5 of 50
Accep
ted
Man
uscr
ipt
5
AD. The breakdown is thought to be due to chronically progressive AD
neuropathology with underlying molecular mechanisms leading downstream to
neuronal and synaptic dysfunction and ultimately to neuronal loss. Such AD-
characteristic structural and functional changes are hypothesized to reflect at least
partially the progressive impairment of fiber tract connectivity and integrity (Meguro
et al., 1999; Morrison and Hof, 2002; Morrison et al., 1986; Stoub et al., 2006),
suggesting that the disconnection in AD is evident at both the functional and structural
level.
The aim of the current review is to characterize neural network changes with regard to
(a) the characteristics of AD-related neuropathology, the distribution within the brain
and association with dementia severity, (b) functional breakdown, both within the
functional and structural domains of the brain, of specific networks associated with
impaired cognitive function such as memory, (c) possible applications to the clinical
domain, and (d) future approaches for understanding the multi-dimensional nature of
network changes and the behavioural and cognitive changes that they produce. The
associations between brain pathology and indices of functional and structural
connectivity may help our understanding of the role of connectivity in brain function.
We will review studies investigating the neuroanatomical spread of AD-related
pathology, and studies using functional magnetic resonance imaging (fMRI) and
electroencephalographic (EEG) data to investigate functional networks, as well as
studies utilizing diffusion tensor imaging (DTI) to investigate structural changes.
2.0 Neuropathology and its spread in the brain
Page 6 of 50
Accep
ted
Man
uscr
ipt
6
Current understanding of the effects of focal damage on neural networks is
rudimentary, even though such understanding could provide greater insight into
important neurological and psychiatric disorders. AD is characterised by chronically
progressive neurodegenerative mechanisms that translate clinically into multi-domain
cognitive decline, complex psychopathological and behavioural disturbances with
subsequent loss of function to perform day-to-day tasks. One key mechanistic
molecular and histopathological hallmark is proposed to relate to intracellular
hyperphosphorylation of micotubuli-associated tau protein, progressive neurofibrillary
changes such as formation of paired helical filaments (PHF) and neurofibrillary
tangles (NFT), dystrophic neurits, and extracellular neuritic plaques (NP) (Braak and
Braak, 1995; Khachaturian, 1985; Mirra et al., 1991). Another major mechanistic
strand is described in the amyloidogenic cascade hypothesis with pathological
cleavage of the amyloid precursor protein (APP), leading to non-neuritic deposition of
fibrillar Aβ and production of toxic oligomers, dimmers and trimers within the brain
regions (Selkoe, 1994). The development of NFT, leading to microstructural
degeneration within the axon and cell body (Grundke-Iqbal et al., 1986), is associated
with neuronal death (Gomez-Isla et al., 1996), and downstream global cognitive
decline (Arriagada et al., 1992b; Berg et al., 1998, Arriagada, 1992 #220) early in the
course of AD.
The location and distribution of AD-related molecular mechanisms and
neuropathological lesions lend support to the hypothesis that AD is in part a
disconnection syndrome characterized by the loss of afferent and efferent connections
of regional allo- and neocortical areas associated with the death of pyramidal neurons
(Morrison and Hof, 2002; Morrison et al., 1986). The earliest regions affected by AD
Page 7 of 50
Accep
ted
Man
uscr
ipt
7
pathology are the transentorhinal cortex, the parahippocampal gyrus and the
hippocampal formation. It has been found that the projections among the
hippocampal formation, entorhinal cortex, and amygdala contained NFT and the ends
of the projections contained amyloid deposition (Hyman et al., 1990). The location of
the neuropathology is such that it affects intracortical property neurons specifically
(Armstrong, 1993; Mann, 1996; Pearson et al., 1985; Van Hoesen et al., 1991). The
NFT predominate in layers III and V of the association areas in the frontal, temporal
and parietal lobes as well as in layers II and IV of the limbic periallocortex. The
pyramidal neurons are located within these layers, enabling cortico-cortical
connections between the cerebral hemispheres. It has been suggested that the
distribution of the NFT across the AD brain exhibits a pattern where the most
vulnerable cortical regions are those that have connections to the ventromedial regions
of the temporal lobe (Arriagada et al., 1992a). Thus AD pathology may spread in a
stepwise manner from the medial temporal lobes through the cortico-cortical
connections (Bancher et al., 1993; Braak et al., 1993; Buckner et al., 2005). The
hypothesized mechanism is that NFT would be present in the body of long cortico-
cortical pyramidal neuronal cells so that there would be a loss of efferent and afferent
connections to the neocortex (Morrison et al., 1986). Thus brain areas that are the
least affected by NFT in the early stages of AD are those that are far removed, in
terms of cortico-cortical connections from the ventromedial temporal lobes (Arnold et
al., 1991).
3.0 fMRI and PET findings: revealing connectivity in functional brain networks
3.1 Cognitive Function Domain
Page 8 of 50
Accep
ted
Man
uscr
ipt
8
There is growing evidence that brain activity to support a cognitive function occurs
within large-scale brain networks rather than within single isolated brain regions. The
volume of studies on brain connectivity between brain regions has increased steadily
since the earliest studies reported about 15 years ago (McIntosh et al., 1994; Murphy
et al., 1993). For the definition of connectivity of brain activity between brain regions,
two major concepts have been applied (Horwitz, 2003). The first concept refers to
functional connectivity, i.e., the correlation between neuronal changes within one
brain region related to another (Friston, 1998). This approach does not allow for a
causal interpretation of the influence among different brain regions, but is purely
correlational in nature. Functional connectivity has been applied to explore the
correlative pattern of brain activity (Bokde et al., 2001; Horwitz et al., 1987). In
contrast, effective connectivity refers to the influence of one brain region onto the
other where that direction of influence can be explicitly modelled, using approaches
such as structural equation modelling (McIntosh and Gonzalez, 1994), autoregressive
correlation, or dynamic causal modelling (Friston et al., 2003) (for review see
(Ramnani et al., 2004)). These approaches are especially well suited to test specific
hypotheses about the function of a particular neuronal network (McIntosh et al.,
1994).
Another approach to investigate the networks activated by a task is demonstrated by
Sperling and colleagues who utilized independent component analysis (ICA) to
examine the networks that activate or de-activate during an associate memory task
(Celone et al., 2006). The memory network across groups included the visual areas,
hippocampus, bilateral dorsolateral prefrontal cortex and posterior parietal cortices
supporting the hypothesis that a specific large-scale network underpins associative
Page 9 of 50
Accep
ted
Man
uscr
ipt
9
encoding. The study included healthy controls, mildly cognitive impaired subjects
(MCI) and AD patients, and they found a continuum of activation in the hippocampus
from healthy controls to hyperactivation in more mildly impaired MCI subjects to
hypoactivation in more severe MCI subjects to no activation in AD patients. The
nonlinear changes in activation in the network across the various groups provided
further evidence of an initial study suggesting this non-linear dynamic in the
hippocampus (Dickerson et al., 2005). A study using resting measures of glucose
metabolism found medial temporal lobe hypometabolism to be associated with
memory encoding impairments, but the study also found significant correlations to
parietal-temporal association cortices and frontal areas that may be part of a
compensatory process in the AD patients (Desgranges et al., 1998). Decreased
activation of the medial temporal lobe was also found in an encoding task in both
MCI subjects and AD patients compared to healthy controls (Machulda et al., 2003).
To investigate the early changes in networks is not only possible through the
examination of brain activation changes but also by examination of the transition
phase between activation and rest (Fox et al., 2005a). Rombouts and colleagues
(Rombouts et al., 2005b) found that the transition phase between blocks of an
encoding task and fixation led to significant differences between healthy controls,
MCI subjects and AD patients over a network of regions that included the medial
temporal lobe areas as well as visual processing areas, and frontal cortices. Even
though the first indications of AD neuropathology may be present in the medial
temporal lobe, it affects a significant number of regions outside of this initial area due
to the high interconnectivity of the brain.
Page 10 of 50
Accep
ted
Man
uscr
ipt
10
In AD, working memory is also significantly impaired at an early stage of the clinical
manifestation of the disease. Patients with AD are severely impaired in a delay
response task of visuo-spatial memory (Simone and Baylis, 1997; Stuart-Hamilton et
al., 1988) and both visual and visuospatial short-term memory are impaired even in
predementia subjects with MCI (Alescio-Lautier et al., 2007), where impaired
visuospatial processing contributes significantly to deficits in every day skill in AD
(Perry and Hodges, 2000). fMRI and PET-based studies showed that impaired visual
working memory correlated with brain activity within the posterior parietal
association cortex, prefrontal cortex, and thalamus (Collette et al., 1997; Desgranges
et al., 1998, Collette, 1997 #2) in AD. Only few studies have examined changes in
network-related changes in activity in relation to visual working memory breakdown
in AD. Functional connectivity analysis of PET-data showed that patients with AD
exhibit, in comparison to elderly healthy controls, reduced functional connectivity
between the prefrontal cortex and hippocampus and the prefrontal-occipital areas
during a delayed-matched to sample task of face stimuli (Grady et al., 1993). The
reduced connectivity between the prefrontal cortex and visual occipital areas was
consistent with findings for a perceptual matched-to-sample task of face stimuli in AD
(Horwitz et al., 1995). In another study examining a delay-match-to sample- task with
different duration of the delay interval, the age-matched healthy controls showed
increased activity in the bilateral prefrontal and parietal cortex with increasing delay,
whereas the patients had increased activity in the right prefrontal, anterior cingulate
and left amygdala. Task performance in both groups was correlated with the right
prefrontal cortex with the addition that performance in the AD patients was also
correlated to the left amygdala (Grady et al., 2001). Taking the right prefrontal cortex
as reference for functional connectivity analysis, Grady and colleagues found that in
Page 11 of 50
Accep
ted
Man
uscr
ipt
11
healthy controls there was strong functional connectivity to a network of other frontal
areas and posterior cortex regions while the AD patients had strong functional
connectivity only to other frontal regions. It was found that the left amygdala in the
AD patients had strong functional connectivity to the left prefrontal cortex and other
posterior brain regions whereas the healthy controls has strong functional connectivity
only to other posterior cortices. Grady and colleagues suggested that the results show
a functional disconnection between the hippocampus and the frontal cortices in the
AD patients, and that the disconnection was underlying the memory deficit in the AD
patients. A further study in AD patients demonstrated that the recruitment of
additional regions in the prefrontal cortex in AD patients was correlated with
performance in a semantic and episodic memory task while the healthy controls
utilized a different network with the networks correlated to task performance (Grady
et al., 2003).
Insert Figure 1 near here
Not only are memory networks affected as shown by another study that examined a
the functional connectivity between the fusiform gyrus and a wide cortical network
across the brain (see Figure 1) (Bokde et al., 2006b). In this study, the task was to
decide if two faces presented simultaneously were identical, and the reference region
for the functional connectivity was the right fusiform gyrus, a key region in the
perception of faces. Of interest in this study was that the activation in this task was
not altered between the MCI subjects and the healthy controls (Bokde et al., 2008),
suggesting that connectivity within a network is first altered due to the putative AD
neuropathology and then changes in activation occur in the brain. It may be that
Page 12 of 50
Accep
ted
Man
uscr
ipt
12
before recruitment of compensatory regions for a cognitive task, functional
connectivity would be the first step leading to increased activation in a region that
would activate as a compensatory mechanism. These issues would have to be
examined within a longitudinal study framework to be able to answer these questions
in more detail.
Effective connectivity can also be quantified using electrophysiological measures of
brain activation (Astolfi et al., 2005; Massimini et al., 2005; Moran et al., 2008;
Ursino et al., 2007) and initial work has been done in AD (see review (Uhlhaas and
Singer, 2006)). Connectivity in resting state EEG was increased in the theta and delta
bands and it was associated with a decrease in power in the alpha and beta bands
(Babiloni et al., 2006; Jelles et al., 2008). The changes across the brain are not only
frequency specific but also vary according to the spatial location and reflect the
remaining connectivity pattern in the AD brain (Stam et al., 2007; Stam et al., 2003).
There have been few studies using EEG with cognitive paradigms in AD and one
study found reduced synchronization in the alpha and beta bands during the delay
phase (maintenance) of a working memory task (Pijnenburg et al., 2004).
One of the best delineated neuronal networks in humans is the visual system of the
human brain. The ventral pathway has been thought to underlie object identification
whereas the dorsal pathway has been associated with processing of the spatial location
of objects (Haxby et al., 1991; Ungerleider and Mishkin, 1982). Analysis of effective
connectivity assessed by structural equation modelling showed evidence for the
correlated activity within each pathway (McIntosh et al., 1994) specific to object vs
location matching. Such results of connectivity analyses were central to identify an
important feature: connectivity analysis supports the notion of distinct functionally
Page 13 of 50
Accep
ted
Man
uscr
ipt
13
integrated networks. For example McIntosh and colleagues (McIntosh and Gonzalez,
1994) showed that the differences in effective connectivity of the ventral and dorsal
visual pathways, as well as the inter-hemispheric effective connectivity, were
different as a function of the cognitive task performed by the healthy elderly subjects.
This approach has been applied to investigating the changes in healthy aging (Della-
Maggiore et al., 2000) and in AD patients (Horwitz et al., 1995). In the study by
Horwitz and colleagues they found that there was a functional disconnection between
the dorsolateral prefrontal cortex and regions in the occipital-temporal lobes. The AD
patients as a compensatory process for the disconnection recruited additional regions
in the frontal lobes (Horwitz et al., 1995).
3.2 Coherent Resting Networks in the Brain
Recent developments on the functional and structural organization of the brain have
demonstrated that there are large-scale networks across the brain that are defined
through a coherent low frequency signal (Damoiseaux et al., 2006; Fox et al., 2005b).
This spatial temporal structure extends throughout the brain and has been found also
in non-human primates (Vincent et al., 2007). The findings of Vincent and colleagues
(Vincent et al., 2007) suggest that fluctuations of spontaneous activity across
anatomically interconnected brain regions constitute a fundamental principle of brain
organization. Such an interpretation is supported by the fact that organized patterns of
brain activity are present in both humans and non-human primates. The resting
networks have generated new issues when examining brain activation due to a
cognitive task, such the relationship between the task-associated network and the
resting networks in the brain (Buckner and Vincent, 2007; Greicius and Menon,
2004).
Page 14 of 50
Accep
ted
Man
uscr
ipt
14
The most investigated network among the spontaneous fluctuation networks that has
been investigated is the default mode network (DMN) which is of particular interest
for AD research because it includes the medial temporal lobes and the posterior
cingulate – two key areas supporting memory function as well as affected very early
in the disease – as well as lateral inferior parietal cortex and medial frontal areas. It is
hypothesized that the DMN is active when a person does not do a goal oriented task,
and it is hypothesized to mediate awareness of the internal state of the person as well
as awareness of the external environment surrounding the subject (Gusnard et al.,
2001; Raichle et al., 2001). The DMN is deactivated (suppressed) during performance
of a cognitive task and it has been measured using two approaches: (a) comparing a
cognitive task to rest condition and examining the regions deactivated during the task,
and (b) analysis using only resting fMRI datasets to measure the DMN. In a study
with young and old healthy controls and AD patients performing a semantic
classification task (Lustig et al., 2003), it was found that the deactivation in lateral
parietal regions was similar in all three groups, while the medial frontal areas showed
it was reduced between young and old healthy controls with no further reduction in
the AD group. The medial parietal region and posterior cingulate showed decreased
deactivation between young and old, with much less deactivation in the AD group
compared to both groups. Further examination of the temporal profile of the
activation in this medial parietal region/posterior cingulate found that the healthy
subjects deactivated this region during the task but that the AD patients had a constant
level of activation across the semantic task and the control task. Another study that
examined the deactivation during a cognitive task (Rombouts et al., 2005a), in this
case a visual encoding and a working memory task in MCI, AD and healthy controls,
Page 15 of 50
Accep
ted
Man
uscr
ipt
15
found that the deactivation in the medial frontal areas discriminated between the HC
and the MCI (medial frontal) and AD (anterior cingulate) groups. In addition, the
precuneus significantly discriminated between the healthy controls and the MCI and
AD groups.
When utilizing only resting state fMRI data sets, Greicius and colleagues (Greicius et
al., 2004) found significant differences in the DMN between AD and healthy controls
in the hippocampus and posterior cingulate region. Another study found decreased
functional connectivity between the right hippocampus in AD patients to cortical
regions across the brain compared to healthy controls (Wang et al., 2006). In
particular, the regions with disrupted connectivity comprised in part the DMN
showing further support to the network-related nature of brain function disruption.
3.3 Interaction among networks
Given that the DMN is deactivated during a cognitive task, it is of high interest to
examine if there is an interaction between the task-related network activated and the
DMN. In a study using associate encoding, Celone and colleagues (Celone et al.,
2006) found that the deactivation in the lateral and medial parietal regions was
reciprocally related to the activation in the task related activation in the hippocampus.
Across the 3 groups in the study, the activation in the hippocampus (bilaterally) was
strongly inversely linearly correlated to the deactivation in the bilateral parietal
regions. Further evidence that an impaired deactivation contributes to impairment in
the task related activation and task performance is a study in AD patients that found a
linear correlation between increased activation in medial temporal areas in the patient
group during an associative memory paradigm to the impaired deactivation in the
Page 16 of 50
Accep
ted
Man
uscr
ipt
16
parietal areas (Pihlajamaki et al., 2008). In AD patients the level of deactivation in the
medial parietal areas was correlated with memory performance with less deactivation
correlated with less successful encoding. Further support for the role of the DMN in
cognitive tasks was found in a visual perception study with MCI subjects who showed
decreased negative functional connectivity between a visual perception area and the
medial frontal areas of the DMN compared to healthy controls (Bokde et al., 2006a).
Thus initial evidence from functional imaging studies indicate that not only do
networks mediate cognitive function but also that the interactions among networks,
among them the default network, have a linear association with performance. The role
of the default network in cognition and how it might underpin it is unresolved.
3.4 Connectivity Dysfunction due to Changes in White Matter
White matter lesions (WML) are prevalent in AD with about one-third of autopsy-
confirmed cases of AD (Mirra et al., 1991) but are frequently found in ageing as well
(Erkinjuntti et al., 1994; Scheltens et al., 1992). WML including microstructural
changes may be related to factors such as microvascular damage leading to
hypoperfusion and white matter degeneration (Bailey and Kandel, 1993; de la Torre,
2004). The significance of white matter lesions alone for cognitive decline is not
clear. This may be partially explained by the fact that for the assessment of
macrostructural white matter changes, including presence of lacunae and white matter
hyperintensities, lesion ratings were often averaged across large brain areas in
previous studies, thus compromising the sensitivity to detect a correlation between the
white matter changes and the decline in specific cognitive functions (Burns et al.,
2005; Snowdon et al., 1997). Importantly, conventional T2-weighted MRI is sensitive
Page 17 of 50
Accep
ted
Man
uscr
ipt
17
towards macrostructural lesions but less so for the assessment of microstructural white
matter changes that could remain undetected in so-called normal appearing white
matter areas (Bozzali et al., 2001). Such microstructural changes that are common in
AD, however, can be detected with diffusion tensor imaging (DTI).
Damage of the membrane and degeneration of intra-axonal microtubules are
associated with neurofibrillary changes that may lead to axonal microstructural
damage (Grundke-Iqbal et al., 1986). DTI is sensitive for the detection of
microstructural alterations, even though it is not clear which specific intra-axonal
changes lead to changes in the DTI-assessed diffusivity (Beaulieu, 2002). The
reduction of cellular integrity of nerve fibres as observed in AD may result in less
constrained motion of the water molecules and thus higher diffusion and lower
anisotropy values (Basser and Jones, 2002). If the integrity of neuronatomical
connectivity between brain regions is an important determinant of neuronal network
activity, damage to neuronal connections within a network should have a specific
impact on the effective connectivity within the network, as often found in
neurodegenerative diseases (Au Duong et al., 2005a; Au Duong et al., 2005b; Grady
et al., 2001).
The apparent diffusion coefficient (ADC) can be calculated per voxel, where the
intensity value is proportional to the diffusion of protons. Differences in the spatial
orientation of diffusivity are expressed by fractional anisotropy (FA). Based on the
assumption that the diffusivity is maximal in the direction of fibre tracts, the voxel-
by-voxel determination of FA can be used in order to tract fibres at the macroscopic
level (Mori and van Zijl, 2002). A number of DTI-based studies in AD patients have
Page 18 of 50
Accep
ted
Man
uscr
ipt
18
demonstrated decreased FA and increased diffusivity in temporal lobes of the brain, a
key area in AD. ROI-based DTI studies have shown increased ADC in patients with
mild to moderate AD within the temporal stem (Hanyu et al., 1998; Kantarci et al.,
2001), anterior and posterior cingulate gyrus (Kantarci et al., 2001; Rose et al., 2000;
Takahashi et al., 2002; Zhang et al., 2007), and the corpus callosum (Bozzali et al.,
2002; Duan et al., 2006; Sydykova et al., 2006; Teipel et al., 2007b; Xie et al., 2006).
Insert Figure 2 near here
Using a multivariate factor analysis approach to analyze the FA maps obtained from
AD patients and healthy controls, Teipel and colleagues (Teipel et al., 2007b) found
that there was a spatially correlated pattern of decreased FA in intracortical fibers that
included key tracts in the temporal lobes (see Figure 2). The intracortical fibres with
lower FA in the AD group included the anterior corpus callosum, white matter of the
parahippocampal gyrus and fornix, left fasciculus longitudinal inferior, white matter
areas in left inferior and middle temporal gyri, white matter areas in bilateral frontal
lobes, right posterior cingulate and right middle occipital gyrus. The decreased FA in
these areas is consistent with previous findings of grey matter structural changes, such
as decreased FA in the parahippocampal white matter would indicate
neurodegeneration of the white matter fibers that connect to the allocortical areas of
the temporal lobes, which are affected early in AD (Price et al., 2001). The FA
decreases in the fornix probably correspond to the loss of neurons in the
hippocampus, as the fibers from the hippocampus project via the fornix to the
mamillary bodies. Thus it can be seen that the decreased FA values occur within
white matter fibres that connect to medial temporal areas. It is consistent with the
Page 19 of 50
Accep
ted
Man
uscr
ipt
19
hypothesis that the changes produced by AD neuropathology within the brain follow a
network of connections that arise from the medial temporal areas (Arriagada et al.,
1992a; Morrison and Hof, 2002; Morrison et al., 1986). The association cortices
located in the parietal lobes have long cortical-cortical connection to the medial
temporal areas and these areas in the parietal lobes are also one of the first areas
affected by AD. Furthermore, the decline in the corpus callosum, whose fibers
connect both hemispheres, is consistent with atrophy of these areas due to grey matter
declines on the cortex (Hampel et al., 2000; Hensel et al., 2002; Teipel et al., 2003;
Teipel et al., 2002; Teipel et al., 1998; Teipel et al., 1999). It was found that left
cingulum fibers, which connect the anterior thalamus, the cortical cingulum, and the
association cortices in the frontal, temporal and parietal cortices and the hippocampus
to each other, were correlated with free recall, verbal recognition and Boston Naming
test performance in AD patients (Fellgiebel et al., 2008). The various regions that the
cingulum fibers connect have been shown to be involved in the various tasks of
memory such as encoding (in the hippocampus), retrieval and recognition (in the
posterior cingulate, the retrosplenial cortex, and posterior and medial parietal cortex).
4.0 Future Perspectives of Research in Connectivity in AD
The studies reviewed here suggest that AD is in part a disorder caused by
disconnection within cognitive networks and the failure of the brain to integrate the
functionality of the various regions into an effective and efficient network. The
assessment of the integrity of specific fibre tracts can be used in order to assess its
association with the degree of functional brain activity (Toosy et al., 2004). If the
integrity of neuroanatomical connectivity between brain regions is an important
Page 20 of 50
Accep
ted
Man
uscr
ipt
20
determinant of neuronal network activity, damage to neuronal connections within a
network should have a specific impact on the effective connectivity within the
network (Au Duong et al., 2005a; Au Duong et al., 2005b; Grady et al., 2001). One
approach to investigate the effects of structural changes on function would be to
integrate effective connectivity and structural connectivity measures together using a
multiple regression technique. It would allow testing for associations between the
various connectivity measures and task performance. Given the initial changes
within the temporal lobes and the memory domain in AD, a possible starting point for
integration of function and structure would be in memory tasks and a neural network
that would include the temporal lobes (see Figure 3).
Insert Figure 3 near here
The proposed integration of effective connectivity can also be done using EEG-based
measures of effectivity connectivity and DTI. Thus one study examined the changes
in the resting-state EEG inter-hemispheric connectivity between MCI subjects and
healthy controls and changes in diffusivity across the brain (Teipel et al., 2008). It was
found that the temporal-parietal coherence in the alpha band was correlated with FA
and MD values in the white matter in posterior regions of the brain in both MCI and
HC. In the frontal lobes coherence in the alpha band was correlated with diffusivity
in the frontal lobes, anterior corpus callosum, and thalamus only in the MCI group.
Thus, they showed an association between inter-hemispheric coherence changes and
alterations in the alpha and beta bands of resting-state EEG.
Page 21 of 50
Accep
ted
Man
uscr
ipt
21
In addition, methodological advancements in time series analysis would allow for a
more detailed understanding of the changes in the phase-related changes in the fMRI-
related signal. In addition, the coherent resting networks will provide another avenue
for investigating neural networks and their breakdown, as well as plasticity processes,
as from the available evidence it seems that the resting networks do not have
compensatory processes, at least, as would be manifested by the recruitment of other
regions. Thus the task-related networks and the resting coherent networks seem to
have different properties and the compensatory processes are different between these
networks. The interaction of networks and the dynamics of the resting coherent
networks is a rich area of current research.
Future studies should also examine the multi-modal nature of networks, both the
structural and functional components that define a network. Given the large changes
that the brain undergoes with the presence of AD-related neuropathology, the changes
will manifest themselves not only in the functional and structural domains but also in
how the changes in the two domains interact with one another. For example, one
study examined how brain activation in the fusiform gyrus during a perceptual task
was dependent upon grey matter density along the ventral and dorsal visual pathways
(Teipel et al., 2007a). Thus not only local grey matter atrophy may influence
activation, but also atrophic changes at other nodes of the network. These issues need
to be further investigated.
5.0 Clinical Applications
Assuming that cognition requires a high level of interaction among regions of a
network, it may be that alterations in the interaction among these regions may be the
Page 22 of 50
Accep
ted
Man
uscr
ipt
22
first biological indicator of active cellular molecular mechanisms and related
neuropathology in the AD brain and that changes in connectivity would be followed
by changes in activation. A multi-modal approach to investigating neural networks
would inform the sequence of events that would lead to a breakdown of cognitive
function at the earliest stages of the disease process. These issues have not been
investigated and may play a critical role in the development of disease modifying
compounds for AD.
Early detection of AD is of dramatically increasing importance since many new
compounds claiming disease-modifying effects are currently being tested in phase 2
and 3 clinical trials. These drug candidates for secondary AD prevention would
preferentially be investigated in patients during earlier presymptomatic stages since it
is hypothesized that these compounds would be more effective when less damage to
the brain has occurred. With this objective in mind, a recent position paper (Dubois
et al., 2007) proposed that the research criteria for the diagnosis of AD should be
refined and updated to be able to detect the earliest clinical stages of AD. The new
criteria would be centred on clinically significant deficits in episodic memory and an
abnormal measure on one or more biomarkers among structural MRI, molecular PET
imaging, or cerebrospinal fluid amyloid beta or tau protein analysis. The greater use
of biological based information such as the implementation of mechanistic biological
markers of action (MoA) is particularly important in order to reflect safety and
outcome induced by mechanistically active and potentially disease modifying
compounds, such as amyloid lowering agents, amyloid immunization strategies,
gamma and beta secretase inhibitors, or approaches targeting inflammation, oxidative
stress or tau hyperphosphorylation and tangle formation. Understanding network
Page 23 of 50
Accep
ted
Man
uscr
ipt
23
changes as early as possible within the chronically progressive AD disease course
holds promise to provide an effective indirect means for early presymptomatic
detection of AD pathology and to help create enriched and stratified early target
populations for presymptmatic trials, as the disease modifying strategies could be
potentially preventive of further progression to irreversible damage to brain structure
and function. This notion is currently strongly supported by regulatory authorities,
such as the FDA and the EMEA, searching for more suitable drug trial designs and
biological safety and outcome measures in the development of therapies in the field of
neurodegeneration.
The proposed integration of both structural and functional connectivity, as illustrated
in Figure 3, would help increase our understanding of the underlying biological
processes in AD but also could be applicable to investigation of early prodromal
stages of AD. In the medium to long term perspective, a specific application of the
proposed methods would serve as an effective and dynamic tool for enrichment of
pre-symptomatic treatment trials so that the inclusionary criteria for the trial are more
specific to AD. The proposed approach could also be implemented as a secondary
outcome variable in phase 3 confirmatory clinical trials in which the outcome does not
depend upon specific neuropathology or mechanisms in the brain.
6.0 Conclusions
Experimental data across a wide variety of approaches suggest that connectivity plays
a critical role in mediating cognitive function and that the breakdown of connectivity,
both in the functional and structural system domain, plays a major role in the
development of AD. One critical element in understanding the disconnection
Page 24 of 50
Accep
ted
Man
uscr
ipt
24
hypothesis in AD is that the spreading in the brain of the AD-related neuropathology
as disease severity increases is through the large cortical-cortical pyramidal neurons.
The location of the AD-related neuropathology in the mild stage of the disease is in
the medial temporal lobes. Thus one sees the development of AD-related
neuropathology in very specific regions of the cortex that are structurally connected
while adjacent regions remain free of AD neuropathology. In addition, the studies
using structural and functional imaging methodologies showed that networks
mediated cognitive performance and their breakdown was correlated to decreased
cognitive performance. In particular, compensatory networks in patients were shown
to linearly correlate to cognitive performance. Thus the various approaches to
understanding networks could be valuable tools in developing new approaches for
early diagnosis of AD and for predicting the effectivity of possible treatment
strategies.
Acknowledgements
The authors acknowledge support for their research from the Volkswagen Foundation
(Germany), the German Ministry of Education and Research (BMBF), the German
Brain Foundation (Hirnliga), the European Union’s FP7 and Social Funds
Programmes, the Alzheimer’s Association (USA), Science Foundation Ireland (SFI),
the Health Research Board (Ireland), and the Health Service Executive (HSE,
Ireland).
Page 25 of 50
Accep
ted
Man
uscr
ipt
25
Figure 1. Map of the regions showing statistically significant differences in the linear
correlation coefficient between healthy control and MCI groups. Figure from (Bokde
et al., 2006a).
Figure 2. Projection of the positive and negative components of the canonical image
into voxel space—3D-reconstruction. The canonical image in voxel space projected
on a 3D-reconstruction of the T1-weighted template brain. A block has been cut-out
from the anterior right hemisphere, opening the view on the internal capsule in height
of the central sulcus at Talairach–Tournoux y-coordinate − 17. Red to yellow:
components of the canonical images that are reduced in AD relative to controls. Blue
to green: components of the canonical images that are increased in AD relative to
controls. Figure from (Teipel et al., 2007b)
Figure 3. Integration of fMRI and DTI results into a picture of network connectivity
over the two domains. The image on the upper left illustrates a hypothesized network,
on the right side tehre are two images of activation, where the activation occurs in
areas indicated by the network. The other images show some of the tracts that
connect the various regions on the network.
Page 26 of 50
Accep
ted
Man
uscr
ipt
26
References
Alescio-Lautier B, Michel BF, Herrera C, Elahmadi A, Chambon C, Touzet C, et al.
Visual and visuospatial short-term memory in mild cognitive impairment and
Alzheimer disease: role of attention. Neuropsychologia 2007; 45: 1948-60.
Armstrong RA. Is the clustering of neurofibrillary tangles in Alzheimer's patients
related to the cells of origin of specific cortico-cortical projections? Neurosci
Lett 1993; 160: 57-60.
Arnold SE, Hyman BT, Flory J, Damasio AR, Van Hoesen GW. The topographical
and neuroanatomical distribution of neurofibrillary tangles and neuritic
plaques in the cerebral cortex of patients with Alzheimer's disease. Cereb
Cortex 1991; 1: 103-16.
Arriagada PV, Growdon JH, Hedley-Whyte ET, Hyman BT. Neurofibrillary tangles
but not senile plaques parallel duration and severity of Alzheimer's disease.
Neurology 1992a; 42: 631-9.
Page 27 of 50
Accep
ted
Man
uscr
ipt
27
Arriagada PV, Marzloff K, Hyman BT. Distribution of Alzheimer-type pathologic
changes in nondemented elderly individuals matches the pattern in
Alzheimer's disease. Neurology 1992b; 42: 1681-8.
Astolfi L, Cincotti F, Babiloni C, Carducci F, Basilisco A, Rossini PM, et al.
Estimation of the cortical connectivity by high-resolution EEG and structural
equation modeling: simulations and application to finger tapping data. IEEE
Trans Biomed Eng 2005; 52: 757-68.
Au Duong MV, Audoin B, Boulanouar K, Ibarrola D, Malikova I, Confort-Gouny S,
et al. Altered functional connectivity related to white matter changes inside the
working memory network at the very early stage of MS. J Cereb Blood Flow
Metab 2005a; 25: 1245-53.
Au Duong MV, Boulanouar K, Audoin B, Treseras S, Ibarrola D, Malikova I, et al.
Modulation of effective connectivity inside the working memory network in
patients at the earliest stage of multiple sclerosis. Neuroimage 2005b; 24: 533-
8.
Page 28 of 50
Accep
ted
Man
uscr
ipt
28
Babiloni C, Ferri R, Binetti G, Cassarino A, Dal Forno G, Ercolani M, et al. Fronto-
parietal coupling of brain rhythms in mild cognitive impairment: a
multicentric EEG study. Brain Res Bull 2006; 69: 63-73.
Bailey CH, Kandel ER. Structural changes accompanying memory storage. Annu Rev
Physiol 1993; 55: 397-426.
Bancher C, Braak H, Fischer P, Jellinger KA. Neuropathological staging of Alzheimer
lesions and intellectual status in Alzheimer's and Parkinson's disease patients.
Neurosci Lett 1993; 162: 179-82.
Basser PJ, Jones DK. Diffusion-tensor MRI: theory, experimental design and data
analysis - a technical review. NMR Biomed 2002; 15: 456-67.
Beaulieu C. The basis of anisotropic water diffusion in the nervous system - a
technical review. NMR Biomed 2002; 15: 435-55.
Berg L, McKeel DW, Jr., Miller JP, Storandt M, Rubin EH, Morris JC, et al.
Clinicopathologic studies in cognitively healthy aging and Alzheimer's
disease: relation of histologic markers to dementia severity, age, sex, and
apolipoprotein E genotype. Arch Neurol 1998; 55: 326-35.
Page 29 of 50
Accep
ted
Man
uscr
ipt
29
Bokde AL, Lopez-Bayo P, Meindl T, Pechler S, Born C, Faltraco F, et al. Functional
connectivity of the fusiform gyrus during a face-matching task in subjects with
mild cognitive impairment. Brain 2006a; 129: 1113-24.
Bokde AL, Tagamets MA, Friedman RB, Horwitz B. Functional interactions of the
inferior frontal cortex during the processing of words and word-like stimuli.
Neuron 2001; 30: 609-17.
Bokde ALW, Lopez-Bayo P, Born C, Dong W, Meindl T, Leinsinger G, et al.
Functional Abnormalities of the Visual Processing System in Subjects with
Mild Cognitive Impairment: an fMRI study. Psychiatr Res: Neuroimaging
2008; 163: 248-59.
Bokde ALW, Lopez-Bayo P, Meindl T, Pechler S, Born C, Faltraco F, et al.
Functional connectivity of the Fusiform Gyrus during a face matching task in
subjects with mild cognitive impairment. Brain 2006b; 129: 1113-24.
Bozzali M, Falini A, Franceschi M, Cercignani M, Zuffi M, Scotti G, et al. White
matter damage in Alzheimer's disease assessed in vivo using diffusion tensor
magnetic resonance imaging. J Neurol Neurosurg Psychiatry 2002; 72: 742-6.
Page 30 of 50
Accep
ted
Man
uscr
ipt
30
Bozzali M, Franceschi M, Falini A, Pontesilli S, Cercignani M, Magnani G, et al.
Quantification of tissue damage in AD using diffusion tensor and
magnetization transfer MRI. Neurology 2001; 57: 1135-7.
Braak H, Braak E. Staging of Alzheimer's disease-related neurofibrillary changes.
Neurobiol Aging 1995; 16: 271-8; discussion 278-84.
Braak H, Braak E, Bohl J. Staging of Alzheimer-related cortical destruction. Eur
Neurol 1993; 33: 403-8.
Buchel C, Coull JT, Friston KJ. The predictive value of changes in effective
connectivity for human learning. Science 1999; 283: 1538-41.
Buckner RL, Snyder AZ, Shannon BJ, LaRossa G, Sachs R, Fotenos AF, et al.
Molecular, structural, and functional characterization of Alzheimer's disease:
evidence for a relationship between default activity, amyloid, and memory. J
Neurosci 2005; 25: 7709-17.
Buckner RL, Vincent JL. Unrest at rest: Default activity and spontaneous network
correlations. Neuroimage 2007.
Page 31 of 50
Accep
ted
Man
uscr
ipt
31
Burns JM, Church JA, Johnson DK, Xiong C, Marcus D, Fotenos AF, et al. White
matter lesions are prevalent but differentially related with cognition in aging
and early Alzheimer disease. Arch Neurol 2005; 62: 1870-6.
Celone KA, Calhoun VD, Dickerson BC, Atri A, Chua EF, Miller SL, et al.
Alterations in Memory Networks in Mild Cognitive Impairment and
Alzheimer's Disease: An Independent Component Analysis. J. Neurosci. 2006;
26: 10222-10231.
Collette F, Salmon E, Van der Linden M, Degueldre C, Franck G. Functional anatomy
of verbal and visuospatial span tasks in Alzheimer's disease. Hum Brain Mapp
1997; 5: 110-8.
Damoiseaux JS, Rombouts SA, Barkhof F, Scheltens P, Stam CJ, Smith SM, et al.
Consistent resting-state networks across healthy subjects. Proc Natl Acad Sci
U S A 2006; 103: 13848-53.
de la Torre JC. Is Alzheimer's disease a neurodegenerative or a vascular disorder?
Data, dogma, and dialectics. Lancet Neurol 2004; 3: 184-90.
Page 32 of 50
Accep
ted
Man
uscr
ipt
32
Della-Maggiore V, Sekuler AB, Grady CL, Bennett PJ, Sekuler R, McIntosh AR.
Corticolimbic interactions associated with performance on a short-term
memory task are modified by age. J Neurosci 2000; 20: 8410-6.
Desgranges B, Baron JC, de la Sayette V, Petit-Taboue MC, Benali K, Landeau B, et
al. The neural substrates of memory systems impairment in Alzheimer's
disease. A PET study of resting brain glucose utilization. Brain 1998; 121:
611-31.
Dickerson BC, Salat DH, Greve DN, Chua EF, Rand-Giovannetti E, Rentz DM, et al.
Increased hippocampal activation in mild cognitive impairment compared to
normal aging and AD. Neurology 2005; 65: 404-11.
Duan JH, Wang HQ, Xu J, Lin X, Chen SQ, Kang Z, et al. White matter damage of
patients with Alzheimer's disease correlated with the decreased cognitive
function. Surg Radiol Anat 2006; 28: 150-6.
Dubois B, Feldman HH, Jacova C, Dekosky ST, Barberger-Gateau P, Cummings J, et
al. Research criteria for the diagnosis of Alzheimer's disease: revising the
NINCDS-ADRDA criteria. Lancet Neurol 2007; 6: 734-46.
Page 33 of 50
Accep
ted
Man
uscr
ipt
33
Erkinjuntti T, Gao F, Lee DH, Eliasziw M, Merskey H, Hachinski VC. Lack of
difference in brain hyperintensities between patients with early Alzheimer's
disease and control subjects. Arch Neurol 1994; 51: 260-8.
Fellgiebel A, Schermuly I, Gerhard A, Keller I, Albrecht J, Weibrich C, et al.
Functional relevant loss of long association fibre tracts integrity in early
Alzheimer's disease. Neuropsychologia 2008; 46: 1698-1706.
Fox MD, Snyder AZ, Barch DM, Gusnard DA, Raichle ME. Transient BOLD
responses at block transitions. Neuroimage 2005a; 28: 956-66.
Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME. The
human brain is intrinsically organized into dynamic, anticorrelated functional
networks. Proc Natl Acad Sci U S A 2005b; 102: 9673-8.
Friston KJ. The disconnection hypothesis. Schizophr Res 1998; 30: 115-25.
Friston KJ, Harrison L, Penny W. Dynamic causal modelling. Neuroimage 2003; 19:
1273-302.
Gomez-Isla T, Price JL, McKeel DW, Jr., Morris JC, Growdon JH, Hyman BT.
Profound loss of layer II entorhinal cortex neurons occurs in very mild
Alzheimer's disease. J Neurosci 1996; 16: 4491-500.
Page 34 of 50
Accep
ted
Man
uscr
ipt
34
Grady CL, Furey ML, Pietrini P, Horwitz B, Rapoport SI. Altered brain functional
connectivity and impaired short-term memory in Alzheimer's disease. Brain
2001; 124: 739-56.
Grady CL, Haxby JV, Horwitz B, Gillette J, Salerno JA, Gonzalez-Aviles A, et al.
Activation of cerebral blood flow during a visuoperceptual task in patients
with Alzheimer-type dementia. Neurobiol Aging 1993; 14: 35-44.
Grady CL, McIntosh AR, Beig S, Keightley ML, Burian H, Black SE. Evidence from
functional neuroimaging of a compensatory prefrontal network in Alzheimer's
disease. J Neurosci 2003; 23: 986-93.
Greicius MD, Menon V. Default-mode activity during a passive sensory task:
uncoupled from deactivation but impacting activation. J Cogn Neurosci 2004;
16: 1484-92.
Greicius MD, Srivastava G, Reiss AL, Menon V. Default-mode network activity
distinguishes Alzheimer's disease from healthy aging: evidence from
functional MRI. Proc Natl Acad Sci U S A 2004; 101: 4637-42.
Page 35 of 50
Accep
ted
Man
uscr
ipt
35
Greicius MD, Supekar K, Menon V, Dougherty RF. Resting-State Functional
Connectivity Reflects Structural Connectivity in the Default Mode Network.
Cereb Cortex 2008.
Grundke-Iqbal I, Iqbal K, Tung YC, Quinlan M, Wisniewski HM, Binder LI.
Abnormal phosphorylation of the microtubule-associated protein tau (tau) in
Alzheimer cytoskeletal pathology. Proc Natl Acad Sci U S A 1986; 83: 4913-
7.
Gusnard DA, Akbudak E, Shulman GL, Raichle ME. Medial prefrontal cortex and
self-referential mental activity: relation to a default mode of brain function.
Proc Natl Acad Sci U S A 2001; 98: 4259-64.
Hampel H, Teipel SJ, Alexander GE, Horwitz B, Pietrini P, Hippius H, et al. Corpus
callosum measurement as an in vivo indicator for neocortical neuronal
integrity, but not white matter pathology, in Alzheimer's disease. Ann N Y
Acad Sci 2000; 903: 470-6.
Hanyu H, Sakurai H, Iwamoto T, Takasaki M, Shindo H, Abe K. Diffusion-weighted
MR imaging of the hippocampus and temporal white matter in Alzheimer's
disease. J Neurol Sci 1998; 156: 195-200.
Page 36 of 50
Accep
ted
Man
uscr
ipt
36
Haxby JV, Grady CL, Horwitz B, Ungerleider LG, Mishkin M, Carson RE, et al.
Dissociation of object and spatial visual processing pathways in human
extrastriate cortex. Proc Natl Acad Sci U S A 1991; 88: 1621-5.
Hensel A, Wolf H, Kruggel F, Riedel-Heller SG, Nikolaus C, Arendt T, et al.
Morphometry of the corpus callosum in patients with questionable and mild
dementia. J Neurol Neurosurg Psychiatry 2002; 73: 59-61.
Horwitz B. The elusive concept of brain connectivity. Neuroimage 2003; 19: 466-70.
Horwitz B, Grady CL, Schlageter NL, Duara R, Rapoport SI. Intercorrelations of
regional cerebral glucose metabolic rates in Alzheimer's disease. Brain Res
1987; 407: 294-306.
Horwitz B, McIntosh AR, Haxby JV, Furey M, Salerno JA, Schapiro MB, et al.
Network analysis of PET-mapped visual pathways in Alzheimer type
dementia. Neuroreport 1995; 6: 2287-92.
Horwitz B, Warner B, Fitzer J, Tagamets MA, Husain FT, Long TW. Investigating
the neural basis for functional and effective connectivity. Application to fMRI.
Philos Trans R Soc Lond B Biol Sci 2005; 360: 1093-108.
Page 37 of 50
Accep
ted
Man
uscr
ipt
37
Hyman BT, Van Hoesen GW, Damasio AR. Memory-related neural systems in
Alzheimer's disease: an anatomic study. Neurology 1990; 40: 1721-30.
Jelles B, Scheltens P, van der Flier WM, Jonkman EJ, da Silva FH, Stam CJ. Global
dynamical analysis of the EEG in Alzheimer's disease: frequency-specific
changes of functional interactions. Clin Neurophysiol 2008; 119: 837-41.
Kantarci K, Jack CR, Jr., Xu YC, Campeau NG, O'Brien PC, Smith GE, et al. Mild
cognitive impairment and Alzheimer disease: regional diffusivity of water.
Radiology 2001; 219: 101-7.
Khachaturian ZS. Diagnosis of Alzheimer's disease. Arch Neurol 1985; 42: 1097-105.
Lee L, Friston K, Horwitz B. Large-scale neural models and dynamic causal
modelling. Neuroimage 2006; 30: 1243-54.
Luria AR. The working brain. New York: Basic, 1973.
Lustig C, Snyder AZ, Bhakta M, O'Brien KC, McAvoy M, Raichle ME, et al.
Functional deactivations: change with age and dementia of the Alzheimer
type. Proc Natl Acad Sci U S A 2003; 100: 14504-9.
Page 38 of 50
Accep
ted
Man
uscr
ipt
38
Machulda MM, Ward HA, Borowski B, Gunter JL, Cha RH, O'Brien PC, et al.
Comparison of memory fMRI response among normal, MCI, and Alzheimer's
patients. Neurology 2003; 61: 500-6.
Mann DM. Pyramidal nerve cell loss in Alzheimer's disease. Neurodegeneration
1996; 5: 423-7.
Massimini M, Ferrarelli F, Huber R, Esser SK, Singh H, Tononi G. Breakdown of
cortical effective connectivity during sleep. Science 2005; 309: 2228-32.
McIntosh AR. Contexts and catalysts: a resolution of the localization and integration
of function in the brain. Neuroinformatics 2004; 2: 175-82.
McIntosh AR, Gonzalez F. Structural equation modelling and its application to
network analysis in functional brain imaging. Hum Brain Mapp 1994; 2: 2-22.
McIntosh AR, Grady CL, Ungerleider LG, Haxby JV, Rapoport SI, Horwitz B.
Network analysis of cortical visual pathways mapped with PET. J Neurosci
1994; 14: 655-66.
Meguro K, Blaizot X, Kondoh Y, Le Mestric C, Baron JC, Chavoix C. Neocortical
and hippocampal glucose hypometabolism following neurotoxic lesions of the
Page 39 of 50
Accep
ted
Man
uscr
ipt
39
entorhinal and perirhinal cortices in the non-human primate as shown by PET.
Implications for Alzheimer's disease. Brain 1999; 122: 1519-31.
Mirra SS, Heyman A, McKeel D, Sumi SM, Crain BJ, Brownlee LM, et al. The
Consortium to Establish a Registry for Alzheimer's Disease (CERAD). Part II.
Standardization of the neuropathologic assessment of Alzheimer's disease.
Neurology 1991; 41: 479-86.
Moran RJ, Molholm S, Reilly RB, Foxe JJ. Changes in effective connectivity of
human superior parietal lobule under multisensory and unisensory stimulation.
Eur J Neurosci 2008; 27: 2303-12.
Mori S, van Zijl PC. Fiber tracking: principles and strategies - a technical review.
NMR Biomed 2002; 15: 468-80.
Morrison JH, Hof PR. Selective vulnerability of corticocortical and hippocampal
circuits in aging and Alzheimer's disease. Prog Brain Res 2002; 136: 467-86.
Morrison JH, Scherr S, Lewis DA, Campbell MJ, Bloom FE. The laminar and
regional distribution of neocortical somatostatin and neuritic plaques:
implications for Alzheimer's disease as a global neocortical disconnection
Page 40 of 50
Accep
ted
Man
uscr
ipt
40
syndrome. In: Scheibel AB and Weschler AF, editors. The biological
substrates of Alzheimer's disease. New York: Academic Press, 1986: 115-31.
Murphy DG, DeCarli CD, Daly E, Gillette JA, McIntosh AR, Haxby JV, et al.
Volumetric magnetic resonance imaging in men with dementia of the
Alzheimer type: correlations with disease severity. Biol Psychiatry 1993; 34:
612-21.
Pearson RC, Esiri MM, Hiorns RW, Wilcock GK, Powell TP. Anatomical correlates
of the distribution of the pathological changes in the neocortex in Alzheimer
disease. Proc Natl Acad Sci U S A 1985; 82: 4531-4.
Perry RJ, Hodges JR. Relationship between functional and neuropsychological
performance in early Alzheimer disease. Alzheimer Dis Assoc Disord 2000;
14: 1-10.
Pihlajamaki M, DePeau KM, Blacker D, Sperling RA. Impaired medial temporal
repetition suppression is related to failure of parietal deactivation in Alzheimer
disease. Am J Geriatr Psychiatry 2008; 16: 283-92.
Pijnenburg YA, v d Made Y, van Cappellen van Walsum AM, Knol DL, Scheltens P,
Stam CJ. EEG synchronization likelihood in mild cognitive impairment and
Page 41 of 50
Accep
ted
Man
uscr
ipt
41
Alzheimer's disease during a working memory task. Clin Neurophysiol 2004;
115: 1332-9.
Price JL, Ko AI, Wade MJ, Tsou SK, McKeel DW, Morris JC. Neuron number in the
entorhinal cortex and CA1 in preclinical Alzheimer disease. Arch Neurol
2001; 58: 1395-402.
Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL. A
default mode of brain function. Proc Natl Acad Sci U S A 2001; 98: 676-82.
Ramnani N, Behrens TE, Penny W, Matthews PM. New approaches for exploring
anatomical and functional connectivity in the human brain. Biol Psychiatry
2004; 56: 613-9.
Rogers BP, Morgan VL, Newton AT, Gore JC. Assessing functional connectivity in
the human brain by fMRI. Magn Reson Imaging 2007.
Rombouts SA, Barkhof F, Goekoop R, Stam CJ, Scheltens P. Altered resting state
networks in mild cognitive impairment and mild Alzheimer's disease: An
fMRI study. Hum Brain Mapp 2005a; 26: 231-9.
Page 42 of 50
Accep
ted
Man
uscr
ipt
42
Rombouts SA, Goekoop R, Stam CJ, Barkhof F, Scheltens P. Delayed rather than
decreased BOLD response as a marker for early Alzheimer's disease.
Neuroimage 2005b; 26: 1078-85.
Rose SE, Chen F, Chalk JB, Zelaya FO, Strugnell WE, Benson M, et al. Loss of
connectivity in Alzheimer's disease: an evaluation of white matter tract
integrity with colour coded MR diffusion tensor imaging. J Neurol Neurosurg
Psychiatry 2000; 69: 528-30.
Scheltens P, Barkhof F, Valk J, Algra PR, van der Hoop RG, Nauta J, et al. White
matter lesions on magnetic resonance imaging in clinically diagnosed
Alzheimer's disease. Evidence for heterogeneity. Brain 1992; 115 ( Pt 3): 735-
48.
Selkoe DJ. Cell biology of the amyloid beta-protein precursor and the mechanism of
Alzheimer's disease. Annu Rev Cell Biol 1994; 10: 373-403.
Simone PM, Baylis GC. The role of attention in a spatial memory task in Alzheimer
disease patients. Alzheimer Dis Assoc Disord 1997; 11: 140-52.
Page 43 of 50
Accep
ted
Man
uscr
ipt
43
Snowdon DA, Greiner LH, Mortimer JA, Riley KP, Greiner PA, Markesbery WR.
Brain infarction and the clinical expression of Alzheimer disease. The Nun
Study. Jama 1997; 277: 813-7.
Stam CJ, Jones BF, Nolte G, Breakspear M, Scheltens P. Small-world networks and
functional connectivity in Alzheimer's disease. Cereb Cortex 2007; 17: 92-9.
Stam CJ, van der Made Y, Pijnenburg YA, Scheltens P. EEG synchronization in mild
cognitive impairment and Alzheimer's disease. Acta Neurol Scand 2003; 108:
90-6.
Stoub TR, deToledo-Morrell L, Stebbins GT, Leurgans S, Bennett DA, Shah RC.
Hippocampal disconnection contributes to memory dysfunction in individuals
at risk for Alzheimer's disease. Proc Natl Acad Sci U S A 2006; 103: 10041-5.
Stuart-Hamilton IA, Rabbitt PM, Huddy A. The role of selective attention in the
visuo-spatial memory of patients suffering from dementia of the Alzheimer
type. Compr Gerontol [B] 1988; 2: 129-34.
Sydykova D, Stahl R, Dietrich O, Ewers M, Reiser MF, Schoenberg SO, et al. Fiber
Connections between the Cerebral Cortex and the Corpus Callosum in
Page 44 of 50
Accep
ted
Man
uscr
ipt
44
Alzheimer's Disease: A Diffusion Tensor Imaging and Voxel-Based
Morphometry Study. Cereb Cortex 2006.
Takahashi S, Yonezawa H, Takahashi J, Kudo M, Inoue T, Tohgi H. Selective
reduction of diffusion anisotropy in white matter of Alzheimer disease brains
measured by 3.0 Tesla magnetic resonance imaging. Neurosci Lett 2002; 332:
45-8.
Teipel SJ, Bayer W, Alexander GE, Bokde AL, Zebuhr Y, Teichberg D, et al.
Regional pattern of hippocampus and corpus callosum atrophy in Alzheimer's
disease in relation to dementia severity: evidence for early neocortical
degeneration. Neurobiol Aging 2003; 24: 85-94.
Teipel SJ, Bayer W, Alexander GE, Zebuhr Y, Teichberg D, Kulic L, et al.
Progression of corpus callosum atrophy in Alzheimer disease. Arch Neurol
2002; 59: 243-8.
Teipel SJ, Bokde AL, Born C, Meindl T, Reiser M, Moller HJ, et al. Morphological
substrate of face matching in healthy ageing and mild cognitive impairment: a
combined MRI-fMRI study. Brain 2007a; 130: 1745-58.
Page 45 of 50
Accep
ted
Man
uscr
ipt
45
Teipel SJ, Hampel H, Alexander GE, Schapiro MB, Horwitz B, Teichberg D, et al.
Dissociation between corpus callosum atrophy and white matter pathology in
Alzheimer's disease. Neurology 1998; 51: 1381-5.
Teipel SJ, Hampel H, Pietrini P, Alexander GE, Horwitz B, Daley E, et al. Region-
specific corpus callosum atrophy correlates with the regional pattern of
cortical glucose metabolism in Alzheimer disease. Arch Neurol 1999; 56: 467-
73.
Teipel SJ, Pogarell O, Meindl T, Dietrich O, Sydykova D, Hunklinger U, et al.
Regional networks underlying interhemispheric connectivity: An EEG and
DTI study in healthy ageing and amnestic mild cognitive impairment. Hum
Brain Mapp 2008.
Teipel SJ, Stahl R, Dietrich O, Schoenberg SO, Perneczky R, Bokde AL, et al.
Multivariate network analysis of fiber tract integrity in Alzheimer's disease.
Neuroimage 2007b; 34: 985-95.
Toosy AT, Ciccarelli O, Parker GJM, Wheeler-Kingshott CAM, Miller DH,
Thompson AJ. Characterizing function-structure relationships in the human
Page 46 of 50
Accep
ted
Man
uscr
ipt
46
visual system with functional MRI and diffusion tensor imaging. Neuroimage
2004; 21: 1452-1463.
Uhlhaas PJ, Singer W. Neural synchrony in brain disorders: relevance for cognitive
dysfunctions and pathophysiology. Neuron 2006; 52: 155-68.
Ungerleider LG, Mishkin M. Two Visual Systems. In: Ingle DJ, Goodale MA and
Mansfield RJW, editors. Analysis of Visual Behavior. Cambridge: MIT Press,
1982: 549-86.
Ursino M, Zavaglia M, Astolfi L, Babiloni F. Use of a neural mass model for the
analysis of effective connectivity among cortical regions based on high
resolution EEG recordings. Biol Cybern 2007; 96: 351-65.
Van Hoesen GW, Hyman BT, Damasio AR. Entorhinal cortex pathology in
Alzheimer's disease. Hippocampus 1991; 1: 1-8.
Vincent JL, Patel GH, Fox MD, Snyder AZ, Baker JT, Van Essen DC, et al. Intrinsic
functional architecture in the anaesthetized monkey brain. Nature 2007; 447:
83-6.
Page 47 of 50
Accep
ted
Man
uscr
ipt
47
Wang L, Zang Y, He Y, Liang M, Zhang X, Tian L, et al. Changes in hippocampal
connectivity in the early stages of Alzheimer's disease: evidence from resting
state fMRI. Neuroimage 2006; 31: 496-504.
Xie S, Xiao JX, Gong GL, Zang YF, Wang YH, Wu HK, et al. Voxel-based detection
of white matter abnormalities in mild Alzheimer disease. Neurology 2006; 66:
1845-9.
Zhang Y, Schuff N, Jahng GH, Bayne W, Mori S, Schad L, et al. Diffusion tensor
imaging of cingulum fibers in mild cognitive impairment and Alzheimer
disease. Neurology 2007; 68: 13-9.
Page 48 of 50
Accep
ted
Man
uscr
ipt
Figure_1
Page 49 of 50
Accep
ted
Man
uscr
ipt
Figure_2
Page 50 of 50
Accep
ted
Man
uscr
ipt
Figure_3